A Comparative Study on Tolerance Analysis Approaches

International Symposium on Robust Design 2014, Copenhagen, Denmark B. Schleich, N. Anwer, Z. Zhu, L. Qiao, L. Mathieu, S. Wartzack

A Comparative Study on TA Approaches Outline  Motivation – Geometric Variations Management as a Branch of Robust Design  Proprietary Tolerance Analysis Approach  The Concept of Skin Model Shapes and the Tolerance Analysis based thereon  Case Studies: GD&T Standards and Assembly Sequence  Conclusion and Outlook

© LEHRSTUHL FÜR KONSTRUKTIONSTECHNIK Friedrich-Alexander-Universität Erlangen-Nürnberg Prof. Dr.-Ing. Sandro Wartzack

disturbance

dependability

six sigma

mean shift functional requirements

geometric variations management

concept design

variation reliability robustness noise factor process capability SORM simulation availability tolerance design quality sensitivity analysis FORM

insensitivity taguchi

Schleich 2014/08 Name 05.32.001

robust design methodology safety

2

Motivation Geometric Variations Management as a Part of RD Robust Design Methodology Robust Design Methodology is understood as systematic efforts to achieve insensitivity to noise factors. These efforts are founded on an awareness of variation and can be applied in all stages of product design.

Non-robust Design

Robust Design

Product variation Design

noise factor availability

six sigma

geometric variations management

concept design disturbance Manufacturing and Inspection and dependability Assembly Quality Control reliability robustness

FORM

insensitivity taguchi

Geometric Variations Management mean shift robust design methodology safety functional requirements

process capability SORM simulation sensitivity analysis tolerance design

Schleich 2014/08 Name 05.32.001

Geometric Requirements Tolerance Specification quality Tolerance Synthesis Tolerance Analysis

Ref.: Arvidsson&Gremyr2008, Reuter2000 © LEHRSTUHL FÜR KONSTRUKTIONSTECHNIK Friedrich-Alexander-Universität Erlangen-Nürnberg Prof. Dr.-Ing. Sandro Wartzack

3

Proprietary CAT software Tolerance Analysis Approach Definition of the assembly CAD models and specification of tolerance types and values as well as definition of their individual distributions. 1.

Definition of the assembly sequence (moves), the part/features relative positioning and the mating conditions (e. g. planar or cylindrical). 2.

Specification of Key Characteristics (KCs) and geometric functional requirements, such as gaps or clearances.

3.

Proprietary Tolerance Simulation Tool

Tolerance Representation for CAT

Tolerance Assignment in CAD

pos A par A ft

Definition of Tolerances

Definition of Assembly 2

1

1

Specification of (F)KCs

2

50

1. 34,1% fb

A

2. 27,8% 3. 19,5%

Result Visualization

Tolerance Simulation

4. 12%

Workflow

Information Transfer

Schleich 2014/08 Name 05.32.001

5. Analysis of the outcome data and identification of the main contributors to evaluate their sensitivity to the KCs and the tolerance design robustness. This step is supported by visualization techniques, such as histograms or KC plots.

Consideration only of rotational and translational feature defects!

4.

Simulation of the effect of part tolerances on KCs using a worst-case or statistical approach (methods such as Monte Carlo simulation are used) employing a tolerance simulation model.

Ref.: Prisco & Giorleo 2002, Shah et al. 2007, Mazur et al. 2011, Clozel et al. 2012 © LEHRSTUHL FÜR KONSTRUKTIONSTECHNIK Friedrich-Alexander-Universität Erlangen-Nürnberg Prof. Dr.-Ing. Sandro Wartzack

4

Skin Model Shapes Fundamentals and Concept GeoSpelling defines a Specification as a… …Condition on a

 Intrinsic Characteristic  Situation Characteristics…

CHARACTERISTIC defined from

GEOMETRIC FEATURES created from

by  Partition, Filtration, Extraction, …

Operations

SKIN Model

Schleich 2014/08 Name 05.32.001

The Skin Model Shapes  Skin ModelModel of the physical between the  interface Skin Model is an Infinite workpiece and its Model  environment No Possibility for  Exists in the mind of the Identification or Simulation product developer Translation and  Can be used to represent Operationalization to a the partModel in the analysis & Finite allocation of tolerances

Nominal Model

Skin Model

Skin Model Shapes

Ref.: 2007Mathieu&Ballu, 2008Dantan,Ballu&Mathieu, 2011Ballu © LEHRSTUHL FÜR KONSTRUKTIONSTECHNIK Friedrich-Alexander-Universität Erlangen-Nürnberg Prof. Dr.-Ing. Sandro Wartzack

5

Skin Model Shapes Tolerance Analysis Approach Skin Model Shape Generation Prediction Stage: Modelling of systematic and random deviations Observation Stage: Results from Simulations or Measurement Data

Pre-Processing Nominal Model

Skin Model Shape Generation

Skin Model Shapes

SMS Preparation Partition of SMS using GeoSpelling operations

Skin Model Shape Preparation Assembly Modelling

2

1

Definition of the Assembly Process

Assembly Simulation Model

Relative Positioning

Assembly Position

1. 34,1%

Assembly Process Definition

Schleich 2014/08 Name 05.32.001

Define the assembly process and the assembly sequence

2. 27,8% 3. 19,5%

Comparison for Conformance

Contact Quality Evaluation

4. 12%

Post-Processing

Relative Positioning Employ constrained registration approaches for the relative positioning of Skin Model Shapes © LEHRSTUHL FÜR KONSTRUKTIONSTECHNIK Friedrich-Alexander-Universität Erlangen-Nürnberg Prof. Dr.-Ing. Sandro Wartzack

6

Case Study 1 Consideration of GD&T Standards Quick View Considered Geometric Deviations:  Geometric Deviations are visible on the Block  Flatness, Parallelism & Position Tolerances  ft=fb=0.05, par=0.1, pos=0.2  Gaussian Distribution (Six Sigma)

 No Deviations of the Plates Assembly Sequence:  Straight 3-Point Moves 1. Block on first Plate 2. Plate on Block B A

pos A par A ft

C D

E

Schleich 2014/08 Name 05.32.001

E’

β

G

F B’

A’

α

D ’ F’

C’ 10 50 G’

40 80 © LEHRSTUHL FÜR KONSTRUKTIONSTECHNIK Friedrich-Alexander-Universität Erlangen-Nürnberg Prof. Dr.-Ing. Sandro Wartzack

50 fb

A Ref.: 2001Anselmetti&Mathieu

7

Case Study 1 Consideration of GD&T Standards – Results

Findings Mean Shift between the proprietary CAT tool and the approach based on SMS for dimensional FKCs

Schleich 2014/08 Name 05.32.001

Prop. CAT tool overestimates the effects for the tilt angles

© LEHRSTUHL FÜR KONSTRUKTIONSTECHNIK Friedrich-Alexander-Universität Erlangen-Nürnberg Prof. Dr.-Ing. Sandro Wartzack

8

Case Study 2 Consideration of the Assembly Sequence Quick View per C ft

Considered Geometric Deviations:  Geometric Deviations are visible on the first part (grey) and on the second part (blue)

s

 Flatness, Perpendicularity & Position Tolerances  ft=fb=0.05, per=0.2, pos(A|B)=1.0, pos(C|D)=0.4  Gaussian Distribution (Six Sigma)

D

Assembly Sequence:  Scenario 1:

pos C D ft

 Primary contact (3 points) in the y-direction, secondary contact (2 points) in the x-direction

 Scenario 2:

ft

C

 Primary contact (3 points) in the x-direction, secondary contact (2 points) in the y-direction

30

per A ft

y

B

pos A B ft

x 1

40

Schleich 2014/08 Name 05.32.001

ft

A

© LEHRSTUHL FÜR KONSTRUKTIONSTECHNIK Friedrich-Alexander-Universität Erlangen-Nürnberg Prof. Dr.-Ing. Sandro Wartzack

2

1

2

Ref.: 2001Anselmetti&Mathieu

9

Case Study 2 Consideration of the Assembly Sequence – Results Findings Mean Shift between the proprietary CAT tool and the approach based on SMS for dimensional FKCs

Schleich 2014/08 Name 05.32.001

Little Effect of the Assembly Sequence on the minimal Gap s

© LEHRSTUHL FÜR KONSTRUKTIONSTECHNIK Friedrich-Alexander-Universität Erlangen-Nürnberg Prof. Dr.-Ing. Sandro Wartzack

10

Conclusion and Outlook General Remarks and Further Research Challenges Conclusion Qualitative comparison of the tolerance analysis approaches employing proprietary CAT tools and based on Skin Model Shapes Quantitative comparison of the approaches with two case studies Slight differences between the approaches can be observed Algorithms implemented in proprietary CAT tools are a black box

Outlook Integration of results obtained from computer aided manufacturing tools Consideration of various physical phenomena, such as gravity and friction

Schleich 2014/08 Name 05.32.001

In-depth analysis of the effects of various deviation parameters

Correlation length: 1 © LEHRSTUHL FÜR KONSTRUKTIONSTECHNIK Friedrich-Alexander-Universität Erlangen-Nürnberg Prof. Dr.-Ing. Sandro Wartzack

Correlation length: 5

Correlation length: 10 11

Summer School Tolerance Management FAU Erlangen-Nürnberg, September 2015 2nd Summer School Tolerance Management at the KTmfk  Exchange between Industry, Research and Students in Tolerance Management  Three Day Programme with Industry Cases, Elevator-Pitches and Workshops  September 2015 at the Chair of Engineering Design KTmfk, Friedrich-AlexanderUniversity Erlangen-Nürnberg

Topics:  Methods and Tools for virtual Geometric Variations Management & RD  Recent Trends in Research and Industry

Further Information:

Schleich 2014/08 Name 05.32.001

 http://www.mfk.fau.de/toleranzen

© LEHRSTUHL FÜR KONSTRUKTIONSTECHNIK Friedrich-Alexander-Universität Erlangen-Nürnberg Prof. Dr.-Ing. Sandro Wartzack

12